How would a robot understand its environment and make sense of semantics in the environment to better collaborate with a human world? How can multiple robots use the knowledge of these abstractions to communicate with each other and achieve a mission with minimal cost? These are a few of the questions that inspire the MSM team. This a cross-discipline group where we do full-stack robot research on topics ranging from perception to behavior learning all the way to multi-robot coordination.
Ongoing Research themes:
- Context-based semantic mapping
- Target-driven visual navigation
- Multi-robot exploration and planning
- Optimal path planning in ocean environments
Team Members:
- Vikas Dhiman
- Anwesan Pal
- Yiding (Cassie) Qiu
- Akanimoh Adeleye
- Michelle Sit
Selected Publications:
- Looking at the right stuff: Guided semantic-gaze for autonomous driving. In: Proc. of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE/CVF Seattle, USA, 2020.
- DEDUCE: Diverse scEne Detection methods in Unseen Challenging Environments. In: International Conference on Intelligent Robots and Systems, IEEE/RSJ Macau, 2019.
- WaveToFly: WaveToFly: Control a UAV using Body Gestures. In: Intl. Conf. Robotics and Automation, IEEE Montreal, 2019.
- Coordinating multi-robot systems through environment partitioning for adaptive informative sampling. In: Intl. Conf. Robotics and Automation, IEEE Montreal, 2019.
- On-line Coordination Ŧask for Multi-robot Systemsusing Adaptive Informative Sampling. In: Intl. Symp. Exp. Robotics, IFRR Springer, Buenos Aires, 2018.